This chapter discusses how structural conditions shape the labour market transitions and career trajectories of mid-career and older workers in Belgium. It first looks at removing structural barriers to mid-career job mobility in the Belgian labour market. This includes the role of employment protection legislation and seniority wages, the strong prevalence of age discrimination, and limited regional mobility. The second part of the chapter focusses on limiting transitions out of work at older ages. It considers how the design of work incentives can prevent labour market outflows and early exits into retirement. The chapter also emphasises the importance of preventing labour market exit due to illness at older ages, by strengthening return to work policy through earlier and more effective intervention.
Promoting Better Career Mobility for Longer Working Lives in Belgium
Abstract
Key findings
Structural labour market features shape the behaviour of mid-to-late career workers, including job mobility and transitions into and out of work. Effective policy action needs to take a two‑pronged focus: the aim should be to reduce structural barriers to job mobility, but also to limit flows out of the labour market.
Several structural barriers limit job mobility among mid-career and older workers in Belgium. The level of employment protection legislation is one of the highest in the OECD, with strong tenure‑based differentiations. This can decrease incentives for hiring and moving jobs at older ages. The government is already pursuing reforms to limit the increase of notice periods with tenure, though these will only have noticeable effects in 2043.
Seniority wages are prominent in many collective agreements and limit incentives to hire or retain older workers. Seniority tends to be broadly defined, covering not only experience within a specific firm but also general work experience and, in some cases, even periods of unemployment or incapacity. As a result, incentives to hire or retain older workers are reduced. Seniority wages also reinforce labour market inequality, as they mainly apply for white collar workers.
Age discrimination is a major barrier for job mobility at older ages in the Belgian labour market. There are some promising examples of initiatives to promote non-discriminatory hiring practices and working environments in Belgium, such as the Diversity Plan programme in Brussels. In addition, Wallonia and Brussels offer subsidies for the employment of older workers. However, evaluation results cast serious doubt on the efficiency of such subsidies.
Increasing regional mobility can contribute to improved labour market matching. Co‑operation between the Belgian regional Public Employment Services is increasing, but barriers remain. These include informational asymmetries, limited interoperability of IT systems and frameworks, and lack of follow-up on vacancies shared with other regions. Six out of ten jobseekers in Wallonia and Brussels indicate that they do not know where to find vacancies in Flanders.
Early transitions out of work and into retirement are common among older workers in Belgium. In 2024, the effective retirement age was 61.3 for women and 61.1 for men, substantially below the OECD averages of 63.1 and 64.4. While access to early retirement has been curtailed over time, the Belgian pension system still allows for early retirement without penalties. Recent reforms will introduce bonuses and penalties that are close to actuarially neutral, but only for workers with long career histories. This could distort work incentives and negatively affect women and other workers with unstable careers.
The design of unemployment insurance can shape incentives to move back into work. Financial support is crucial to maintain an adequate income during the job search, but excessively generous support can disincentivise a return to employment. Until recently, the duration of unemployment benefits was unlimited in Belgium, uniquely within the OECD. Since 2026, this has been limited to 24 months. However, workers aged 55 or older with 30 years of career contributions will be exempt, which could reduce work incentives and deepen inequalities between labour market groups.
Work incapacity drives transitions out of work in Belgium, particularly with increasing age. In 2024, the number of people receiving disability benefits surpassed 500 000 for the first time. Belgium has introduced several reforms in recent years to formalise the return-to-work process and encourage earlier intervention. However, the system is complex and return to work still takes place too late. Belgian data shows that only three out of ten return-to-work pathways start within the first six months of sickness absence, while the likelihood of returning to the labour market drops significantly after even three months.
Belgium is undertaking further reforms to its return-to-work policy. Employees will face stronger obligations to participate in return to work, and a first assessment of remaining work capacity will take place within eight weeks of sick leave. However, in the absence of voluntary action, the development of a formal return-to-work plan will only be initiated after six months. Return to work can be initiated informally in the meantime; such informal trajectories appear to play an important role, but data on their take‑up and impact is insufficient. In addition, employers will contribute more to sickness benefit payments, which could increase incentives to invest in prevention and retention measures. However, employers will be exempt from these obligations for workers older than 55.
Partial return to work can be an attractive option for mid-to-late career workers on sick leave. The option for partial return to work has long existed in Belgium, but take‑up is increasing in recent years, with 16.2% of workers in long-term invalidity granted partial authorisation to work in 2023. Evidence on the impact of the scheme on labour market reintegration of workers on sick leave is limited.
Based on these considerations, Belgium should take action to:
1. Incentivise labour market mobility through well-calibrated labour market regulation. Maintain a balanced level of employment protection to encourage mobility while safeguarding job security. Encourage social partners to weaken seniority-based wage increases in collective agreements, with particular focus on the inclusion of non-employment periods and cross-sectoral experience, and lead by example in public sector wage bargaining.
2. Take action against age discrimination in the labour market. Encourage employers to adapt age‑neutral hiring and workplace practices by providing guidance and financial incentives. Strengthen labour market monitoring and research on the prevalence and drivers of discrimination. Move towards phasing out or more strictly targeting wage subsidies for older workers.
3. Continue efforts to strengthen mobility across regions. Invest in improved co‑ordination and information sharing between regional Public Employment Services. Establish an online platform gathering regional labour market information and vacancies across Belgium. Further strengthen investment in language training for jobseekers.
4. Continue reviewing the design of pension and unemployment benefits to strengthen work incentives at older ages. Introduce actuarially neutral bonuses and penalties for all workers. Carefully monitor the impacts of the recent reform of unemployment benefits, with particular focus on the impact of differential provisions based on workers’ career histories.
5. Further reform the return-to-work system to enable earlier and more effective intervention. Redesign return-to-work pathways to introduce the development of return-to-work plans for all workers earlier during sickness absence, with clearly defined timelines and obligations for both employees and employers. Remove exemptions for employer contributions to sickness benefits for older workers, with potential exceptions for newly hired older employees. Strengthen the capacity of return-to-work co‑ordinators, insurance doctors and occupational physicians. Improve data collection and evaluation, including both informal and formal return to work pathways.
Structural features of the labour market play a crucial role in shaping the extent of job mobility in the mid-to-late career. In Belgium, various structural labour market elements constitute barriers to mobility from both the labour supply and labour demand side. These include the level of employment protection and the prevalence of seniority wages, but also widespread age discrimination and limited mobility between regions.
2.1.1. Employment protection needs to be carefully calibrated to maintain job security while encouraging mobility
Employment protection legislation – regulation governing the dismissal of workers – is essential to safeguard job security, share the social cost of dismissals, and incentive investment in workplace skills. In addition, employment protection can help manage structural labour market transformations and smooth transitions, by enabling dismissed workers to search for a job while still employed (Cederlöf et al., 2024[1]).
However, where employment protection legislation is too stringent, it can disincentivise workers from changing jobs and discourage firms from hiring, leading to overall lower levels of labour market dynamism. Across OECD countries, more stringent protection on regular contracts is associated with lower levels of labour market mobility, including both job-to-job mobility and unemployment-to-job mobility (Causa et al., 2022[2]). The policy challenge is therefore to find a well-balanced level of employment protection, avoiding excessive worker turnover and maintaining job security, but also limiting suboptimal levels of mobility (OECD, 2020[3]).
For older workers, very high levels of employment protection can be powerful disincentives to job mobility. Age‑specific provisions are likely to be particularly harmful: they can protect older employed workers from layoffs, but disincentivise hiring and may cause employers to substitute for workers from other age groups (Vodopivec et al., 2019[4]; Behaghel, Crépon and Sédillot, 2008[5]). While most OECD countries have moved away from age‑specific variation in employment protection legislation in past decades, many continue to implement tenure‑based provisions, for instance on severance pay or notice periods (OECD, 2024[6]). Strong tenure‑based increases in employment protection are likely to contribute to lower job mobility of older workers, who are more likely to have long tenures (OECD, 2020[3]).
Belgium has one of the highest levels of employment protection legislation in the OECD (Figure 2.1). One of the driving elements of high levels of employment protection on regular contracts in Belgium are the long notice periods and high levels of severance pay (OECD, 2020[3]). Significantly, provisions on notice periods or severance periods showcase strong tenure‑based features. The notice period or severance payment for individuals with tenure of 9 to 12 months is seven weeks, while it is almost nine time as much (62 weeks) for those with tenure of 20 years. In comparison, on average across OECD countries, notice periods and severance pay are seven times as high at 20 years of job tenure as at nine months of tenure (OECD, 2020[3]).
Given these considerations, there may be some room for loosening employment protection for long-tenured employees to reduce the steepness of tenure‑based differentiations, with a view to encouraging greater labour market mobility at older ages. Belgium is already pursuing reforms to reduce the degree of employment protection. In 2025, the government coalition agreed that for contracts starting in 2026, the maximum notice period will be limited to 52 weeks (starting from 17 years of tenure), instead of increasing to up to 65 weeks, as previously. This provision could be a positive step towards increasing labour market mobility. However, in practice it will have a tangible impact at the earliest in 2043, when the first contracts starting in 2026 will reach tenure of 17 years.
Employment protection legislation (regular contracts), 2019
Source: OECD (2026), Strictness of employment protection (dataset), https://data-explorer.oecd.org/s/3s4 (accessed 16 January 2026).
While Belgian reforms to employment protection legislation are promising, reductions in notice periods need to be carefully calibrated. For instance, evidence from Sweden shows that job search during the notice period is more effective than during unemployment, and that longer notice periods can be associated with lower periods of unemployment and better post-unemployment wages (Cederlöf et al., 2024[1]). As such, it is important that reforms to employment protection are monitored to ascertain impacts on job mobility but also the labour market outcomes of workers following dismissal. In addition, reforms need to be complemented by high-quality active labour market support as well as actions to address additional barriers mid-to-late career workers face in the labour market, notably age discrimination (Section 2.1.3).
2.1.2. Seniority wages for white‑collar jobs discourage the hiring and retention of older workers
Alongside the design of employment protection legislation, the age‑wage structure has a significant influence on labour market mobility in the later career. In Belgium, limited levels of job mobility in the mid-to-late career are driven by the continuing prevalence of seniority wages i.e. the fact that wages increase strongly with the length of employees’ tenure. Seniority-based wage systems can negatively impact the labour market prospects of older workers, and are correlated with lower employment and hiring rates of older workers across OECD countries (D’Addio, Keese and Whitehouse, 2010[7]).
Seniority wages impact the labour market mobility of mid-to-late career workers in several ways. For workers, seniority-based entitlements that are tied to tenure within a specific company or sector can limit incentives to move jobs. For employers, hiring older workers is relatively more costly in the presence of seniority wages, particularly if these do not reflect actual productivity (Eurofound, 2019[8]). Previous research has shown that firms with steeper seniority wage profiles are less likely to hire employees aged 50 or above (Zwick, 2012[9]). In addition, higher seniority wages can negatively affect employers’ efforts to retain older workers. Empirical evidence from Austria suggests that steeper seniority wages lead to earlier labour market exit of older workers and an increased occurrence of “golden handshake” agreements within firms (Frimmel et al., 2018[10]).
Seniority-based wage increases remain common in Belgium. Age‑based salary scales were frequently used in the past, but largely abolished in 2015. However, tenure‑ or experience‑based salary increases remain commonly anchored in collective bargaining agreements. A recent empirical investigation finds that wage pressure by age, defined as the average monthly wage of employees aged 50‑59 relative to those aged 30‑39, is third-highest within a sample of 14 European countries (Minne and Saks, 2023[11]). Equally, based on a qualitative comparison of seniority-based provisions in legislation and collective bargaining agreements across Europe, the prevalence of seniority-based wage increases is relatively pronounced in Belgium (Eurofound, 2019[8]).
However, seniority-based wages do not benefit all workers equally in Belgium. Analyses have shown that seniority wage systems are widespread among white collar workers, as well as in non-commercial sectors (Conseil Central de l’Economie, 2020[12]; Conseil supérieur de l’emploi, 2014[13]). In contrast, for blue‑collar workers, seniority wages apply much less frequently (Adalet McGowan et al., 2020[14]). Wage profiles of Belgian workers showcase that age‑wage curves are much steeper among white‑collar workers (Figure 2.2).
Average gross monthly wages and salaries by age and status, Belgium, 2022
Source: An overview of Belgian wages and salaries, Statbel, https://statbel.fgov.be/en/themes/work-training/wages-and-labourcost/overview-belgian-wages-and-salaries.
It is also notable that the definition of seniority varies widely across collective agreements in Belgium. Seniority may include years of tenure in a specific firm, but also in a similar position across the economy, in the same sector, or even years of experience in general. In many cases, experience is very broadly defined, also taking into account periods of unemployment or inactivity (Conseil Central de l’Economie, 2020[12]). Such increases may not accurately reflect a link between experience and productivity – while research from Belgium shows there are positive productivity returns to tenure, these decrease over time. In this context, strong links between tenure and pay increases imply a progressive delinking of wages from productivity (Gagliardi, Grinza and Rycx, 2022[15]).
Against this background, Belgium should move towards further delinking wages from seniority, rather aligning them as closely as possible with actual productivity. The establishment of age‑neutral wage setting practices can incentivise increased job mobility at older ages and counteract disincentives to hire older labour market participants. This is especially true for the very broad definitions of experience present in some collective agreements in Belgium, which arguably have little relation to actual productivity (Conseil supérieur de l’emploi, 2014[13]). Given the anchoring of seniority wages in collective agreements, social partners must play a crucial role in better aligning wages with productivity (Saks, 2014[16]). However, the government can contribute by advocating for age‑neutral wage setting practices in public sector collective bargaining.
2.1.3. Age discrimination is a major barrier to hiring and career progression of older workers
Even where mid-to-late career workers are looking to move jobs, limited employer demand for hiring them can constitute an impediment to job mobility. Demand side barriers are partially a result of structural labour market features, such as seniority wages. However, beyond these structural factors, age discrimination is a major factor limiting the employment prospects of older workers, creating a need for policy action to tackle age‑biased hiring and workplace policies.
A strong evidence base demonstrates that age discrimination is an impediment to the labour market participation of older workers. A recent meta‑analysis of field experiments on hiring discrimination finds that older workers are 34% less likely to receive a positive response to a job application (Lippens, Vermeiren and Baert, 2023[17]). Age discrimination in hiring is commonly driven by stereotypes against older workers. Recruiters perceive older job candidates to have lower technological skills, flexibility and trainability levels than other applicants (OECD/Generation: You Employed, Inc., 2023[18]; Van Borm, Burn and Baert, 2021[19]).
In Belgium, age discrimination in the labour market is pronounced (UNIA, 2025[20]). A field experiment carried out in Flanders between 2014 and 2015 found that when comparing different combinations of job candidates aged 38, 44 and 50, younger candidates had a 64.3% higher change of receiving an invitation for a job interview, and a 39.7% higher change of getting any positive reaction. (Baert et al., 2016[21]). The study also found that older candidates face less age discrimination when they have more in-field experience, suggesting that discrimination is strongly driven by statistical discrimination i.e. employer perceptions of adverse characteristics of older workers (Baert et al., 2016[21]). A more recent field experiment run in Flanders also presented evidence of age discrimination, showing that applicants aged 56 receive on average 28% fewer positive responses compared to those aged 38, 44 or 50 (Dalle, Lippens and Baert, 2024[22]). In Brussels experimental evidence showed that job applicants aged 50 or older face systematically lower odds of receiving any positive response or an interview invite to a job application (Tobback et al., 2024[23]).
Action is needed to combat age discrimination in hiring and create non-discriminatory working environments
Given the strong prevalence of age discrimination in the Belgian labour market, action by employers and policymakers to invest in age‑neutral hiring processes and working environments is crucial. Such action should particularly target the negative stereotypes employers hold against older workers, which are key drivers of discrimination (Dalle, Lippens and Baert, 2024[22]). With respect to hiring discrimination, employers can take action by revising their recruitment practices to ensure age‑inclusive language, prioritise skills-based hiring and ensure that hiring managers are equipped to evaluate mid-career and older applicants (OECD/Generation: You Employed, Inc., 2023[18]). In addition, broader action should be taken to integrate age‑friendly workplace practices across a range of areas (see discussion in Chapter 3).
Governments can support employers in the development of age‑friendly and non-discriminatory working environments by offering guidance, consulting and potentially financial incentives. In Brussels, the diversity plan programme is an example of how building more inclusive workplaces can be encouraged through policy action (Box 2.1). Similar employer support services related to non-discriminatory working environments could also be implemented in other Belgian regions.
To combat discrimination, including age discrimination, the Brussels region has established the diversity plan (plan de diversité) and its associated diversity label. Interested companies can participate in the programme by putting in place a diversity plan in their company. The diversity plan is developed by a within-company working group with help from a consultant from the Brussels Public Employment Service, ACTIRIS. Companies can apply for financial support from ACTIRIS to implement their diversity plan. Three types of plans are available:
1. A mini-plan, most suitable for companies with 50 employees or less, focussing on a specific issue related to diversity and anti-discrimination. Financial support of EUR 5 000 is available.
2. A Global Plan, a more comprehensive instrument involving a systematic analysis of a company’s diversity and anti-discrimination policy, with structural action across four areas: recruitment and selection, personnel policy, internal communication and external positioning. Co-financing of EUR 10 000 is available.
3. A thematic plan, which allows for in-depth work focussing on a specific challenge. This is most appropriate for companies who have previously completed a global or thematic plan and involves an in-depth thematic analysis focussing on a specific target group or domain. Co-financing of EUR 10 000 is available.
Following presentation of the plan to ACTIRIS and approval by the Brussels Ministry of Labour, a subsidy can be granted. The execution of the diversity plan is subsequently evaluated by ACTIRIS, and, where positive, a diversity label can be awarded to the company. Between 2024 and 2025, 31 companies in Brussels were awarded a diversity label.
Source: Le plan de diversité et le label de diversité, Bruxelles Économie et Emploi, https://economie-emploi.brussels/diversite-plan-label and Choisir un plan de diversité, Actiris, https://www.actiris.brussels/fr/employeurs/choisir-un-plan-de-diversite/.
Moreover, more systematically collecting evidence on age discrimination can contribute to raising awareness of the issue and developing more effective workplace strategies. For instance, between 2021 and 2024, Flanders conducted a baseline measurement of labour market discrimination across sectors, with sectors choosing the research methods and grounds for discrimination investigated (age, ethnicity or gender). The aim of this exercise was to strengthen available information on the extent of discrimination within sectors. Action plans to fight discrimination – including measures such as awareness raising, training for employers and employees and processes to develop objective recruitment processes – have been or are being drawn up based on the results of this research (Department Werk en Sociale Economie, 2024[24]).
Several regions use subsidies to encourage hiring of older workers, but they do not appear to be effective
Subsidies are a popular instrument among policymakers seeking to encourage the hiring of older workers, particularly in the context of persistent age discrimination. Wage subsidies can also be used with the goal of offsetting potential disparities between wages and perceived productivity of older workers, which may be particularly pronounced in countries with strong seniority wage systems such as Belgium (OECD, 2019[25]).
While the evidence on hiring subsidies specifically focussed on older workers is not extensive, existing research raises doubts about their effectiveness (Dalle, Verhofstadt and Baert, 2024[26]; Huttunen, Pirttilä and Uusitalo, 2013[27]; Boockmann, 2015[28]). Subsidies have the potential to increase employment, but they also carry a large risk of deadweight effects, where employers collect subsidies for hiring or employing workers they would have employed anyway. Subsidies for hiring or retaining older workers tend to be associated with very large deadweight effects, with subsidised work frequently replacing unsubsidised work and little net gain in employment of older workers (Boockmann, 2015[28]). In addition, evidence from Belgium suggests that subsidies can reinforce negative stereotypes about the employability of older workers, ultimately contributing to their lower employment prospects (Dalle, Verhofstadt and Baert, 2024[26]).
Permanent wage subsidies for older workers exist in several Belgian regions. In Flanders, reductions in social security contributions for older workers were abolished in July 2024. However, subsidies are still available in both Brussels and Wallonia. In Brussels, the subsidy amounts to EUR 1 000 per quarter for employees aged between 61 and 66 who earn at most EUR 8 323 per quarter. In Wallonia, eligibility conditions for subsidies are more generous, starting at age 55 (57 from April 2026) and applying up to a quarterly income of EUR 18 036. The subsidy amounts to EUR 400 for workers aged between 55 and 57, EUR 1 000 for those aged between 58 and 64, and EUR 1 500 for workers aged 65 or older.
Prior research on Belgian wage subsidies suggests that their contribution to supporting the employment of older workers does not offset the large cost associated with them. A recent evaluation found that the permanent subsidy had some small effects on working time and large positive effects on retention of older workers, but only in sectors where early retirement around age 58 is common, while effects across the overall population are negligible (Albanese and Cockx, 2019[29]). The authors also estimate that the total social cost of the subsidy exceeded the benefits by approximately EUR 2 800 per job created (Albanese and Cockx, 2017[30]).
Overall, wage subsidies tend to be a very costly and frequently ineffective instrument to encourage hiring and retention of older workers, as also demonstrated by evaluations in the Belgian context. Belgium should consider phasing out these instruments, or targeting them much more stringently at more vulnerable labour market groups, such as long-term unemployed older workers. In Flanders, a previous evaluation showed that the abolition of a subsidy targeted at the long-term unemployed led to reduced inflows into employment entries for job seekers aged 45 and above with at least six months of unemployment experience (Desiere, Cabus and Cockx, 2020[31]). Specific support for these vulnerable groups could be particularly valuable in the context of current reforms to reduce the duration of unemployment benefits receipt (see Section 2.2). In line with this more targeted approach, Flanders currently offers a subsidy for hiring individuals who have been non-working jobseekers for at least two years. In other regions, subsidy support is already limited to low-earning older workers, particularly in Brussels. Stricter targeting may serve to increase the effectiveness of the subsidy, though this cannot be definitively claimed without any evaluation (OECD, 2019[25]). However, policies to invest in the employability of older workers (Chapter 4) and create age‑friendly working environments (Chapter 3) are much more likely to effectively support the employment of older workers.
2.1.4. Greater mobility between regions could enable improved labour market matching
Barriers to labour market mobility exist not only across the general labour market, but also between regions. In Belgium, there are stark disparities in employment rates and labour market characteristics across different regions, with Flanders showing overall higher employment rates than Brussels or Wallonia (see Chapter 1). Policies to encourage geographic mobility can be effective where there are significant gaps in unemployment and labour market tightness between regions, and potentially lead to improved employment and earnings prospects for jobseekers (OECD, 2025[32]).
In comparison to other European OECD countries, regional mobility among new hires is already relatively high in Belgium (Causa, Luu and Abendschein, 2021[33]). Reflecting the differences across regional labour markets, Flemish workers are most likely to remain in their region for work, while commuting is more common in Wallonia and Brussels (SPF Emploi, 2022[34]). In 2023, 7.9% of Flemish employees commuted to the Brussels region and 1.0% to the Walloon region. In comparison, 13.8% of all Brussels employees commuted to Flanders and 5.2% to Wallonia, while 4.3% of Walloon residents commuted to Flanders and 9.4% to Brussels (Steunpunt Werk, 2025[35])
Regional labour market mobility is influenced by a range of factors that stretch across many policy areas, including not only labour market policy, but also housing, transport and public services, among others (OECD, 2025[32]; Causa, Abendschein and Cavalleri, 2021[36]). With regard to labour market policies, the Public Employment Service plays an essential role in promoting mobility across regions. The Belgian regional PES are already implementing various actions to improve job mobility across regions and have collaboration agreements which aim to facilitate greater labour market mobility (SPF Emploi, 2022[34]; Carpentier et al., 2023[37]). This includes the exchange of vacancies between the Public Employment Services, the elaboration of joint labour market analyses and examples of various outreach campaigns to employers and jobseekers to encourage regional mobility. The Walloon PES has specific consultants focussed on interregional mobility, while the Brussels PES can make referrals to the Flanders PES for jobseekers expressing willingness to work in a different region. In December 2025, the Walloon and Flemish PES signed a reinforced co‑operation agreement, including for instance joint pilot projects and labour market analyses, as well as automatic data exchange.
There are also promising examples of inter-regional co‑operation in specific sectors. For instance, the Aviato project is an interactive digital platform to facilitate the recruitment and talent retention of employees at Brussels Airport. The project is a joint initiative by the regional PES and a variety of other partners. The platform collates all employment opportunities related to the airport in an online portal and also offers a range of training opportunities.
Despite these promising initiatives, a recent study identified several barriers to more efficient co‑operation between PES to facilitate regional labour mobility (Carpentier et al., 2023[37]). The regional PES in Belgium each operate their own IT systems, which can make co‑ordination difficult, with information exchange taking place on a manual basis in many cases. Vacancies are not always systematically translated. While registration of vacancies in other regions is possible, there is often no follow-up on outcomes.
This lack of co‑ordinated information exchange also affects the ability of jobseekers to navigate the interregional labour market. In a survey analysis, six out of ten jobseekers from Brussels and Wallonia indicated that they do not know where to find vacancies from Flanders (Carpentier et al., 2023[37]). At the same time, seven out of ten jobseekers from Brussels and half of Walloon jobseekers would be willing to work in Flanders, illustrating that there is potential for increasing regional mobility. (Carpentier et al., 2023[37]).
Further efforts to improve co‑ordination and information exchange, also including harmonisation of frameworks and classifications used across regions, would be beneficial to improve regional labour market mobility. This should also include outward-facing harmonisation. Improving information about job opportunities, for instance through job fairs or initiatives to strengthen information exchange and co‑operation between regional PES, can reduce information asymmetries for both employer and jobseekers (OECD, 2025[32]). Ideally, an integrated online portal could collate information on all regional labour markets as a one‑stop shop. Several OECD countries are making efforts in this direction, including for instance the example of the Canadian Job Bank (Box 2.2).
In Canada, the Job Bank, as the official portal of the Canadian Public Employment Service, gathers information on both vacancies and labour market insights across regions. The platform showcases information on vacancies across the country and offers various functions, including job matching and transition tools, as well as career planning advice. The portal also includes information on employment outlooks and wage comparisons by region and occupation, offering jobseekers comparable labour market information across the country. Since 2018, jobseekers can directly indicate their willingness to relocate in their profile.
Job Bank is delivered by Employment and Social Development Canada and relies on collaboration between the provincial and territorial governments. Evaluations by the Canadian PES indicate that the integration of different services in one platform enables jobseekers to acquire distant labour market information and decreases information asymmetries and search costs.
Source: OECD (2025[32]), Addressing Regional Labour Market Imbalances in Austria, https://doi.org/10.1787/0cf6186b-en.
If well-designed, financial support can also play a role in increasing rates of interregional mobility. Between 2022‑2025, there was a financial subsidy for jobseekers unemployed for at least 12 months accepting work in another region, amounting to 25% of unemployment benefits previously received for three months. However, this was abolished as part of the recent government agreement. Evidence suggests that financial support for regional mobility tends to be effective when it is substantial, continuous and targeted at low-educated jobseekers (OECD, 2025[32]).
In addition, it is important to note that language proficiency is a substantial barrier to mobility across regions in Belgium (SPF Emploi, 2022[34]). 75.2% of jobseekers from Brussels and 57.9% of Walloon jobseekers citing limited knowledge of Dutch as a reason not to want to work in Flanders (Carpentier et al., 2023[37]). As such, offering high-quality intensive language training should be a priority for regional PES with the aim of furthering interregional mobility. Employers can also play a role here, by carefully evaluating what knowledge of a specific language is required for a position, as well as offering accessible language training for employees (Carpentier et al., 2023[37]).
Structural barriers and incentives significantly shape job transitions of mid-career and older workers in various ways. To effectively support career transitions and longer labour market participation, policy design needs to focus on two elements: removing structural barriers to job mobility in the labour market, but also limiting transitions out of work. In Belgium, early exits from the labour market have historically been encouraged through policy incentives. Despite past and ongoing reforms, premature labour market exit remains common in international comparison. While early retirement pathways have progressively been restricted, many mid-to-late workers move into inactivity, often linked to work incapacity. Further policy action is urgently needed in Belgium to support and incentivise continued labour market participation at older ages and prevent moves into long-term inactivity or retirement.
2.2.1. Premature transitions out of work have historically been common in Belgium, driven by early retirement schemes
Belgium has a long history of early retirement schemes which have incentivised transitions out of work at older ages. In the 1970s, against a background of economic crises and high youth unemployment, early labour market exit of older workers was encouraged through the introduction of various pathways towards early retirement (Minne and Saks, 2023[11]). In recent decades, Belgium has moved towards encouraging longer labour market participation, making early retirement schemes more restrictive. Pension ages have also been equalised between men and women and increased, and are set to reach 67 in 2030. Nevertheless, early labour market exit remains relatively common relative to other OECD countries, with effective retirement ages significantly below the OECD average (Figure 2.3). In 2022, the effective retirement age was 61.3 for women and 61.1 for men in Belgium, substantially below the OECD average of 63.1 and 64.4.
Effective labour market exit age by gender, 2024
Note: OECD is an unweighted average.
Source: OECD (2025), Pensions at a Glance (dataset), https://data-explorer.oecd.org/s/35v (accessed 17 January 2026).
In the past decades, one of the most prominent pathways towards early retirement in Belgium has been the “employment insurance with company supplement” scheme (formerly pre‑pension). Under this collectively agreed scheme, workers aged 62 with at least 40 years of career contributions who are dismissed can receive generous wage compensation. The benefit is equal to the level of unemployment benefits, topped up by the employer by half the difference between the last net salary and the unemployment benefit. Depending on the collective agreement, more lenient eligibility conditions may apply in certain sectors. Research suggests that the scheme has a significant negative impact on the labour market participation of older workers. Analysis of a 2008 reform that increased the eligibility age for the scheme from 58 to 60 suggests that the lower age threshold is associated with 19 fewer workdays for men and 25.6 fewer workdays for women (De Brouwer and Tojerow, 2024[38]).
In the past, the pre‑pension scheme was extremely popular among older workers in Belgium. However gradual tightening of eligibility conditions has significantly reduced access to the scheme, which is reflected in a sharp reduction in the share of individuals in pension or early retirement between 2005 and 2020 (Figure 2.4). In 2025, the Belgian Government concluded an agreement to definitively end unemployment with company supplement, with the exception of workers with medical issues.
Labour market status by single year of age between 50 and 64 in 2005 and 2020, Belgium
Source: Based on OECD (2024[39]), OECD Economic Surveys: Belgium 2024, https://doi.org/10.1787/c671124e-en, Figure 3.6.
Over the past decades, access to early retirement under the regular pension scheme has also been gradually tightened, with stricter eligibility conditions relating to eligibility ages and career length (Fraikin, Jousten and Lefebvre, 2021[40]; Desmette and Vendramin, 2021[41]). The early retirement age is dependent on the number of years worked, with the earliest possible retirement possible at age 60 for individuals with 44 years of career contributions.
However, Belgium is one of the only countries in the OECD that does not apply any pension penalties in case of early retirement, nor bonuses for individuals who work longer, meaning that individuals who fulfil the conditions for early retirement can receive actuarially unreduced benefits (Fraikin, Jousten and Lefebvre, 2021[40]; OECD, 2025[42]) The current government foresees introducing penalties and bonuses of 5% per year, close to the actuarially neutral rate. This is a step forward in principle, removing disincentives to work beyond the age of eligibility for (early) retirement. However, these bonuses only apply to individuals who have worked for 35 years or longer, and this group would also be exempt from penalties for early retirement. This could undermine the potential of the system to incentivise longer working lives and increase inequality, as large parts of the population, notably women or individuals who have had unstable careers, would be excluded (OECD, 2025[42]).
2.2.2. Reforms to the unemployment insurance system could increase incentives to seek employment for mid-career and older workers
Next to the design of early retirement systems, unemployment insurance can have a clear impact on the job search behaviour and broader labour market outcomes of mid-to-late career workers. A higher level of unemployment benefits can incentivise labour market mobility by reducing the financial risk associated with job transitions (Sjöberg, 2007[43]). For older workers, who may assess mobility as particularly risky, this incentive may be especially strong. More generous benefits can lead to higher levels of job-to-job mobility, but also job quality following re‑employment, by allowing workers to search for a better job match while maintaining an adequate income (Causa et al., 2022[2]).
At the same time, the level and duration of unemployment benefits must be carefully balanced in order to provide sufficient income protection for older workers during their job search while maintaining incentives to return to employment as well as preventing the deterioration of human capital and long periods of non-employment, both of which could lead to a lower quality of subsequent employment. When unemployment insurance is too generous, there is a risk that it can discourage job search and a return to employment, effectively serving as a pathway to retirement for older workers (Causa et al., 2022[2]). Previous OECD research shows that more generous unemployment insurance after one year of unemployment is associated with a lower likelihood of voluntary job mobility, an effect that is particularly large at older ages (OECD, 2024[6]).
In Belgium, the overall generosity of unemployment benefits, in terms of net replacement rates, is higher than the OECD average, though it does not stand out as particularly generous compared to neighbouring countries such as Germany, the Netherlands or France (Figure 2.5, Panel A) (Adalet McGowan et al., 2020[14]). However, Belgium does stand out with respect to the duration of unemployment benefits, which have until recently been time‑unlimited, a unique position relative to other OECD countries (Figure 2.5, Panel B).
In 2025, Belgium implemented a substantial reform of its unemployment insurance system. The most significant change contained in this reform is the limitation of unemployment benefits. Benefit duration will henceforth be capped at 12 months, with workers with long career histories (31 years, gradually increasing to 35 years until 2029) entitled to an additional 12 months, and unemployed individuals moving to social assistance afterward if they are eligible. However, several groups will be exempt from the reform, including notably workers aged 55 or older with at least 30 years of career contributions (to increase to 35 years by 2030). The reform will be gradually phased in until July 2027.
Empirical evidence supports the notion that reforms to decrease the duration of employment benefits are associated with a reduced duration of unemployment for affected workers (Le Barbanchon, 2016[44]; de Groot and van der Klaauw, 2019[45]; Filges, Jonassen and Jørgensen, 2018[46]), with some evidence that these effects are particularly pronounced for older workers (Kyyrä and Pesola, 2020[47]). Conversely, the evidence on the relationship between benefit duration and the quality of the subsequent job match is mixed (Filges, Jonassen and Jørgensen, 2018[46]). While some studies point to a positive relationship between longer benefit duration and the quality of subsequent employment (Kyyrä and Pesola, 2020[47]; Nekoei and Weber, 2017[48]; de Groot and van der Klaauw, 2019[45]), others identify non-significant or even negative effects (Schmieder, von Wachter and Bender, 2016[49]; Le Barbanchon, 2016[44]). In Switzerland, a shortened potential benefit duration was associated with both improved employment prospects and earnings for jobseekers aged 50‑54 (Cottier, Degen and Lalive, 2019[50]).
Hence, the Belgian reform of unemployment benefits, which aligns with previous OECD recommendations (Adalet McGowan et al., 2020[14]), could positively impact transitions into employment by increasing work incentives, including for older workers. However, it is too early to assess the impact of the reform. Its effects will have to be monitored and evaluated carefully, with attention not only to the impact on transitions into employment, but also the quality of job matches. For the reform to effectively translate into transitions into sustainable employment for older workers, it will have to be accompanied by high quality active labour market support (see Chapter 4) and investment in age‑friendly working environments (Chapter 3).
Net unemployment replacement rates by unemployment duration and maximum benefit duration, 2025
Notes: UB: Unemployment benefit; UI: Unemployment insurance; UR: Unemployment rate.
Benefit rules and calculations assume a 40‑year‑old individual in a single household without children and not eligible to any supplement.
Panel A: Net replacement rates give the share of previous net income replaced by unemployment benefits for a single person with previous earnings at the average full-time wage in the 2nd, 6th and 12th month of registered unemployment. Calculations are for a 40‑year‑old single person without children who has been continuously employed since the age of 18 in full-time employment. Rates for a 55‑year‑old are identical except for: Finland (+0.8 p.p.) and Switzerland (+1.8 p.p.) for all periods; Czechia (+31.6 p.p.) and Italy (+1.4 p.p.) for 6 months, and Austria (+4.4 p.p.), Israel (+3.9 p.p.), Italy (+3.2 p.p.) and Japan (+0.2 p.p.) for 12 months. OECD is an unweighted average of the 34 countries shown.
Panel B: For Belgium, the bar stops at 40 months, but the hashes signify the benefit is of unlimited duration. In Canada, the maximum benefit duration is determined by the regional UR. Data are based on the national UR in January 2025. In Japan, UI claimants deemed to be difficult to re‑employ may receive up to 11.8 months of UB. United States data are based on the Michigan UB scheme. In Sweden, once the UI expires the Job Guarantee Scheme becomes applicable and is of unlimited duration.
Source: OECD calculations based on output from the OECD tax-benefit model version 2.7.1. OECD calculator of taxes and benefits, https://www.oecd.org/en/data/tools/oecd-calculator-of-taxes-and-benefits.html (accessed 25 March 2026) and How do countries calculate tax liabilities and social benefit entitlements?, OECD Policy descriptions for 2025, https://www.oecd.org/en/topics/sub-issues/income-support-redistribution-and-work-incentives/how-do-countries-calculate-tax-liabilities-and-social-benefit-entitlements.html.
In addition, the exemption from the regulation for older workers with 30 years of career contributions could distort the effects of the reform. More lenient regulation for workers with long contribution histories is likely to disproportionately benefit well-integrated labour market participants, while penalizing women and other groups with disrupted labour market careers. Carefully monitoring the effects of the reform on vulnerable groups will therefore be critical for its success. It should be noted, however, that the majority of older workers will not be affected by the exemption. 10 586 unemployment benefit recipients aged 55 or over, representing ca. 1 in 5 of these workers, are eligible for the exemption.
Finally, the new unemployment benefits regulation also contains novel elements relating to receipt of benefits following voluntary resignation, the so-called “trampoline benefit”. Whereas workers who resigned from their jobs were previously not entitled to unemployment benefits, the new regulation posits that, once in their life, workers with ca. ten years of work experience who quit their job are entitled to six months of unemployment benefits, or 12 months if undertaking training in a shortage occupation. This is a promising provision which could provide an incentive for job mobility, by increasing financial security for workers during a period of job search. Potential misuse of the provision will have to be monitored but should be limited by the already rather strict eligibility conditions, including the limitation to six months of benefit receipt.
2.2.3. Further reforming the return-to-work system is a major concern amid soaring numbers of older workers in work incapacity
With regard to limiting flows out of employment at older ages, work incapacity has emerged as a major issue in Belgium. In recent years, the number of people in long-term incapacity has increased substantially, partially reflecting the closing down of other avenues for early labour market exit. As a response, the return-to-work system has undergone several rounds of reform in past decades, with the aim of enabling mobility back into employment for workers who experience health shocks. However, issues remain within the system, and further action to incentivise and enable return to work for older workers with health issues is a significant policy challenge.
The number of people in long-term work incapacity has soared in Belgium
In Belgium, work incapacity has become a major concern in recent years, as the number of individuals in long-term sickness absence has increased substantially. The Belgian system distinguishes between primary work incapacity i.e. incapacity of up to one year (incapacité primaire de travail/primaire arbeidsongeschiktheid), after which people move into long-term sickness benefit, or invalidity (invalidité/ invaliditeit). In 2022, the number of individuals receiving long-term sickness benefits surpassed 500 000 for the first time. Older workers are particularly likely to be in work incapacity (Figure 2.6, Panel A). In 2023, ca. 87 000 people aged 50‑54 and 117 000 people aged 55‑59 were in long-term work incapacity.
The increasing numbers of individuals in work incapacity are driven by a number of factors. including demographic changes such as the rising labour force participation of women and older workers and the rising prevalence of work-incapacity related health issues in the population (OECD, 2024[39]; De Brouwer and Tojerow, 2023[51]; Conseil supérieur de l’emploi, 2024[52]). In this context, investment in preventive health, including in the workplace, is essential (see discussion in Chapter 3). However, changes in eligibility rules for benefits and economic incentives likely also play a part in influencing rising work incapacity. The gradual closing down of other labour market exit pathways for older workers described in Section 2.1.1, including the phasing out of pre‑pensions and tightening of unemployment benefits job search restrictions, is likely to have led towards a substitution between social insurance programmes, commonly referred to as the “communicating vessels” effect (De Brouwer and Tojerow, 2023[51]).
Individuals in long-term work incapacity by gender and age and take up of benefits by scheme, Belgium
Note: Early retirement refers to individuals on the pre‑pension/unemployment benefit with company supplement scheme, as well as mise en disponibilité prior to retirement. Panel A shows individuals in long-term work incapacity (invalidity scheme), by gender. Panel B shows individuals main labour market status in a given year (individuals who still have an employment contract are not included in work incapacity figures).
Source: OECD calculations based on data from the Datawarehouse marché du travail et protection sociale, Banque Carrefour de la sécurité sociale (Panel A) and Institut national d’assurance maladie‑invalidité (Panel B).
Figure 2.6 illustrates the correlation between rising numbers of workers in work incapacity and falling numbers making use of other social insurance programmes, including the pre‑pension scheme and unemployment insurance. Against the background of tightening eligibility rules, between 2003 and 2022, the number of people registered as unemployment benefit recipients as their primary labour market status decreased from 565 000 to 272 000. In the same time frame, the number of people benefitting from the pre‑pension early retirement scheme decreased from 105 000 to 30 000. Conversely take‑up of invalidity benefit increased substantially in the same period. While shifts in disability benefit take‑up are driven by a multiplicity of factors and this analysis should not be taken as indicative of causality, it does point to interrelations between social insurance schemes along the lines of the “communicating vessels” effect referred to above.
Return-to-work policies have been reformed numerous times
Amid soaring numbers of people in work incapacity, Belgium has introduced numerous reforms over the past years. The primary focus of these reforms has been incentivizing return to work for workers on sickness absence by formalizing the return-to-work process and improving incentives for employers and workers. Reform measures introduced included the standardisation of different stages of the return-to-work process, increased incentives for part-time work during benefit receipt, increased co‑operation with regional PES to enable retraining or provide job search assistance during sickness absence, and the clarification of the roles of different stakeholders in the return-to-work process (De Brouwer and Tojerow, 2023[51]).
The most significant reform related to return to work was the introduction of standardised reintegration pathways in 2016 (for a detailed description, see Box 2.3). Following this reform, workers or employers can initiate an informal or formal “reintegration trajectory” at any point during sickness absence. The formal reintegration trajectory involves an assessment of remaining work capacity by an occupational physician and the subsequent elaboration of a return-to-work plan, if deemed feasible.
If no reintegration trajectory is initiated, or in cases where workers are no longer employed, return-to-work procedures are automatically initiated by the health insurance fund. Patients on sick leave are required to respond to a questionnaire on their capacity to work after ten weeks of absence, and to respond to invitations by the medical doctor or return-to-work co‑ordinator. Patients may face sanctions if they do not participate. An assessment of remaining work capacity is then made after four months of sickness absence. If return to work is deemed feasible, a meeting with a return-to-work co‑ordinator is planned to develop a return-to-work plan. However, the measures recommended as part of the trajectory remain voluntary. Since 2023, there are mandatory meetings between the worker on sick leave and the health insurance’s medical advisor during the fourth, seventh and eleventh month of absence. Depending on the judgement of the insurance doctor, meetings will also take place afterwards, either every year, after two years or after five years.
In Belgium, there are several pathways (or “trajectories”) for return to work. Return to work is managed through the National Institute for Health Insurance and Invalidity (INAMI/RIZIV), which is responsible for paying sickness benefits, in consultation with health insurance funds, occupational and general physicians, employers and Public Employment Services. Return-to-work trajectories can either be initiated by individual workers or employers or are initiated through the health insurance fund. At any time during work incapacity, an employed individual can initiate a return-to-work procedure either informally or through INAMI’s return-to-work co‑ordinators, advisors who are responsible for managing the return-to-work process in consultation with the health insurance funds medical advisors and other stakeholders. If no contact is voluntarily initiated, a return-to work procedure is automatically initiated.
Return to work initiated through the health insurance fund: If return to work is initiated by the health insurance fund, as a first step, after ten weeks of incapacity, the fund sends all individuals in incapacity a questionnaire to gather relevant information on their condition and potential return to work. At the latest four months into work incapacity, the health insurance fund’s medical advisor then makes an assessment of remaining capacity to work, on the basis of medical records and the completed questionnaire. Four outcomes are possible:
1) Category 1: Return to work is likely at the latest after six months of incapacity
2) Category 2: For medical reasons, return to work is not possible
3) Category 3: Return to work is currently not feasible as the priority is medical treatment
4) Category 4: Return to work is possible after work adaptation or training
If return to work is deemed feasible (Category 4), a meeting with a return-to-work co‑ordinator is planned, which takes place at the latest within the sixth month of incapacity. A meeting with the co‑ordinator is also foreseen if individuals were initially classed as category 1 or 3 but circumstances have changed and work adaptation is needed. The co‑ordinator then develops a detailed plan for return to work. For employed workers, this is done, where possible, in collaboration with the occupational physician to facilitate a return to the original job. For unemployed or self-employed individuals, the plan is developed with the health insurance fund’s medical advisor, as well as the Public Employment Service where appropriate (e.g. to facilitate retraining). If the worker does not attend the meeting with the return-to-work co‑ordinator, they can receive a sanction of 2.5% of their incapacity benefit. However, there is no obligation to follow the return-to-work plan developed by the return-to-work co‑ordinator.
Return to work initiated by the employee or employer: Workers can voluntarily initiate return to work at any time during their sickness absence, through either a formal or informal reintegration trajectory. All Belgian companies are affiliated to external services for prevention and protection at work (for more detail, see Chapter 3), which, among other services, provide occupational physicians that can assist with evaluating work capacity and return to work. After four weeks of absence, the occupational physician contacts the worker on leave (usually by post) to inform them of different options for return to work, including an informal consultation with the occupational physician or the initiation of a formal reintegration trajectory. A mandatory work capacity assessment also has to be conducted by the employer within eight weeks of absence.
The informal return to work pathway involves a “pre-return-to-work visit” with the occupational physician designed to facilitate a return to employment. The employee can request such a visit at any point, and the occupational physician proposes a consultation. The occupational physician can propose work adjustments that could enable a return to work, but the informal pathway does not include the obligatory development of a return-to-work plan.
As an alternative to the informal pathway, formal reintegration trajectories exist. These can be initiated by the employee or employer at any time during incapacity. Since 2026, if a worker is still in work incapacity after six months of absence, the employer is required to initiate a formal return trajectory (even where informal return has been attempted). The occupational physician then invites the worker to a medical evaluation, contacts other relevant stakeholders (e.g. the general physician, the employer) and assesses the place of work to determine whether reintegration in the previous job, potentially with adaptations, is possible. The occupational physician must extend at least three invitations to attend the examination to the worker, but if the worker chooses not to attend, the return-to-work trajectory is terminated.
Following the examination, the occupational physician provides their evaluation within 49 days of the initial request for a reintegration trajectory. The potential outcomes of the evaluation are:
1) Decision A: The worker is temporarily unable to return to their original job, and may perform adapted work in the meantime, with the conditions to be specified by the physician
2) Decision B: The worker is definitively unable to return to work in their original job, but adapted work is possible, with the conditions to be specified by the physician
3) Decision C: For medical reasons, evaluation of the possibility to return to work is currently not possible and the return-to-work trajectory is terminated
Following the evaluation phase, the employer has the obligation to develop a reintegration plan where the occupational physician has determined that adaptations to the workplace or work are needed. This plan is developed in consultation with the worker and occupational physician, and must be finalised within 63 days (decision A) or six months (decision B). If workplace adaptations are not possible, the employer must provide a justification for this. The worker then has 14 days to consider the plan and decide whether to accept or refuse the reintegration plan.
There are different points during the reintegration trajectory where the process may be terminated. If this occurs, the occupational physician notifies the health insurance fund, who can then initiate a general return-to-work trajectory.
Source: Réintégration des travailleurs en incapacité de travail, SPF Emploi – Travail et Concertation sociale, https://emploi.belgique.be/fr/themes/bien-etre-au-travail/la-surveillance-de-la-sante-des-travailleurs/reintegration-des?id=45586 and Reprendre un travail pendant une incapacité de travail à l’initiative du médecin-conseil de votre mutualité, INAMI, https://www.inami.fgov.be/fr/themes/reinsertion-socio-professionnelle/reprendre-un-travail-pendant-une-incapacite-de-travail-a-l-initiative-du-medecin-conseil-de-votre-mutualite.
Further reform of the return-to-work system is needed to enable earlier intervention
Early intervention is a crucial element of policies to support return to work following sickness absence. International evidence demonstrates that the likelihood of returning to work drops dramatically after even three months of sickness absence (OECD, 2015[53]). Similarly, descriptive data from Belgium shows that exits from incapacity towards the labour market are much more likely during the first three months of sickness absence (Figure 2.7). Out of all Belgian workers who exited from primary work incapacity (i.e. the first year of sickness absence) towards the labour market in 2024, 67% did so during the first three months of sickness absence. With increasing duration of sickness absence, exits from incapacity towards the labour market become increasingly unlikely. Evaluations have also shown that for employees who had been absent for longer than six months at the start of the reintegration trajectory, the odds of returning to work with the same or another employer were considerably lower (Boets et al., 2024[54]).
Currently, participation in return-to-work pathways still takes place too late in the majority of cases. Only three out of ten formal reintegration trajectories commence within the first six months of sickness absence, as do only one in ten pathways involving accompaniment or retraining by the Public Employment Service (Conseil supérieur de l’emploi, 2024[52]). Half of the latter trajectories only commence after two years of sickness absence. The fact that a return-to-work pathway is not generally developed within even the first six months of sickness absence presents a fundamental obstacle to effective return-to-work support.
Belgium is already undertaking reforms to enable earlier intervention within the return-to-work system and strengthen obligations for both employers and employees. Through reforms introduced in 2025, the meetings with the return-to-work co‑ordinator and occupational physician that form part of the return-to-work trajectory are mandatory and accompanied by stronger sanctions for repeated non-attendance. In addition, employers will have to undertake a mandatory assessment of remaining work capacity eight weeks into sickness absence (previously up to four months unless workers voluntarily initiated it). In companies with more than 20 employees, the employer will have to initiate a formal work reintegration trajectory for workers with remaining work capacity after six months of absence. In cases where an employee is definitively judged unable to carry out their previous employment, reorientation towards the Public Employment Services will automatically take place. Employees will also be able to ask their employer for preventive workplace adaptations to avoid periods of incapacity, though the employer has no obligation to fulfil this request.
Share of individuals exiting from primary work incapacity by duration, 2019‑2024, Belgium
Note: Primary work incapacity refers to the first year of work incapacity.
Source: OECD calculations based on Institut national d’assurance maladie‑invalidité – Baromètre Retour au Travail.
These recent reforms go in the right direction by bringing forward timelines for work capacity assessment and strengthening obligations to participate in reintegration for both employers and employees, thereby moving towards earlier intervention. However, further action is needed, most notably to ensure that the development of return-to-work plans takes place earlier during sick leave. Though the reforms bring the work capacity assessment forward, a formal return-to-work trajectory is only initiated after six months of sickness absence and only in companies with 20 employees or more, unless voluntarily initiated earlier. The purpose of the four‑month gap between the work capacity assessment and the start of the formal reintegration trajectory is to leave time for an informal trajectory, which some previous analyses have shown to be successful in driving reintegration (Boets et al., 2020[55]). However, as no impact evaluation of informal trajectories exist, the driving factors behind such positive outcomes are unclear. Available data shows that individuals who follow an informal trajectory are more likely to be highly educated, have higher net incomes, tend to be employed full-time rather than part time and work in non-manual jobs (Boets et al., 2020[55]). This suggests that voluntary, informal trajectories are initiated by a selected group of individuals who are more likely to return to employment in the first place. Conversely, the formal return to work trajectory tends to be taken up by individuals who are more vulnerable, have less contact with their employer and are therefore less predisposed to successful return to work in the first place (Boets et al., 2024[54]). For this group, early intervention is essential and the initiation of a formal trajectory after six months of absence is much too late to make a successful return to employment likely. In other OECD countries, such as Denmark, Norway and the Netherlands, the development of return-to-work plans takes place much earlier for all workers on sick leave (Box 2.4).
A number of OECD countries have introduced systems to ensure that remaining work capacity and perspectives for return to work are assessed as early as possible during a sickness spell. In Denmark employers and employees have to discuss return to work and any need for workplace accommodation within the first four weeks of absence. Within eight weeks, the health insurance advisory doctor conducts an assessment of fitness for work, reassessing every four to eight weeks. The doctor can then initiate vocational rehabilitation measures, which patients are obliged to participate in. Persons on sick leave can lose their right to sickness benefit if they do not participate in follow-up efforts during sick leave without reasonable justification. Similarly, in Norway, discussions of return to work to develop a follow-up plan must take place within the first month of absence, and a work capacity assessment within eight weeks. Unless there is a compelling case for full sick leave, partial return to work is the default option. As in Denmark, if employees refuse to participate without reasonable cause, the right to sickness benefits may lapse.
In the Netherlands, a fixed return-to-work track with specific milestones and dates has been defined. Within six weeks of sickness absence, employees need to see a social insurance or occupational physician for a work capacity assessment. An action plan for promoting return to work must be defined within eight weeks, including any potential adjustments to the previous job if needed. A reassessment of work capacity is required every six weeks.
Source: OECD (2023[56]), Disability, Work and Inclusion in Korea: Towards Equitable and Adequate Social Protection for Sick Workers, https://doi.org/10.1787/bf947f82-en.
Crucially, further development of the return-to-work system should be underpinned by strengthened monitoring and evaluation efforts. Currently, available data and evaluation on return to work in Belgium are insufficient to draw definitive conclusions on the success of the return-to-work system. In particular, informal return to work plays an important role in reintegration of employees on sickness absence, but data or evidence on its impact is very limited. In addition, information on the drivers of absences, sectoral differences and characteristics that cause permanent exit is lacking. Building on such a strengthened evidence base, further reform efforts could be made in the future, notably to reduce complexity in the system and simplify the number and design of return-to-work pathways.
Reforms to sickness benefits could strengthen employer incentives to invest in health prevention and return to work
As part of its current reform efforts, Belgium is also strengthening obligations for employers, by increasing employer contributions to sickness benefits. From 2026, employers with more than 50 employees will pay a solidarity contribution of 30% of sickness benefit (equivalent to 18% of gross salary) between the second and fifth month of sickness absence (during the first month of absence, the guaranteed salary period, income replacement is covered by the employer). The extension of the period where employers are liable for (a share of) sickness benefits is a potential positive step and in line with previous OECD recommendations (OECD, 2024[39]). It also responds to previous evaluations of the new reintegration trajectories, which highlighted limited financial incentives for employers to invest in reintegration, as well as limited incentives or sanctions for employees to participate in the process (Boets et al., 2020[55]). Transferring financial liability for sickness to employers can strengthen incentives for investment in workplace health prevention and employee retention (OECD, 2024[39]).
However, employers will be exempted from paying additional sickness benefits for workers aged 55 or above. Such an exemption may serve the purpose of counteracting disincentives to hire employees perceived as more liable to be absent due to illness, such as older workers (Conseil supérieur de l’emploi, 2024[52]). Nevertheless, exempting employers from paying sickness benefits for older workers greatly reduces incentives to invest in health prevention and return to work for this group, and also reinforces negative signals and stereotypes regarding the employability of older workers. As such, this provision risks undermining the potential positive effects of the reform. A balanced solution could be to exempt employers from the additional sickness benefit payment only for newly hired older workers, but not for existing employees. Exemptions could additionally be considered, at least temporarily, for newly hired workers who have previously been on long-term sick leave.
The capacity of support services needs to be strengthened
For effective return-to-work support, close co‑ordination is needed across several actors, including the employer, occupational physician, general physician, health insurance fund and potentially the Public Employment Service. Systems with well-developed return-to-work pathways frequently implement a case management approach with efficient co‑ordination across these actors (OECD, 2015[53]). Belgium established the role of return-to-work co‑ordinators in 2022. Return-to-work co‑ordinators are employed by health insurance funds and responsible for face‑to-face contact with benefit recipients, liaising with the advisory doctor, occupational doctor and PES (OECD, 2024[39]).
Through improving communication and co‑ordination in the return-to-work process and improving contact with benefit recipient, return-to-work co‑ordinators could effectively support return to work following sickness absence (De Brouwer and Tojerow, 2023[51]). Research has shown that face‑to-face contact with a return-to-work co‑ordinator is associated with a reduced duration of sickness absence as well as a higher likelihood of returning to work (Dol et al., 2021[57]).
However, given the large and increasing numbers of people on (long-term) sickness absence, a larger number of return-to-work co‑ordinators is needed to effectively support return to work. While the number of co‑ordinators has increased since the establishment of the position in 2022, the overall number of full-time equivalent position hovers around 100 in Belgium (Figure 2.8). This is arguable not sufficient relative to the large number of people in work incapacity.
Furthermore, implementing the recently undertaken reforms and further investing in effective and earlier intervention will require strengthening the capacity of not only return-to-work co‑ordinators, but also occupational physicians and medical advisors of the sickness insurance funds. According to information provided by INAMI/RIZIV, in January 2026, there were 347 medical advisors working across all funds (307 full-time equivalent). As regards occupational physicians, taking together both within-company prevention services and external services, there were ca. 1 246 occupational physicians active in Belgium at the end of 2024 (FPS Santé publique, 2025[58]). Limited capacity of these professionals is a significant barrier to effective return to work support in Belgium.
Belgium is also undertaking additional efforts to improve co‑ordination and information-sharing across the different actors involved in the return-to-work process. A newly established digital platform, TRIO, will enable consultation between the occupational physician, general physician and advisory physician of the health insurance. The platform will be used to exchange medical and administrative information regarding return to work.
Number of return-to-work co‑ordinators, 2022‑2025, Belgium
Note: Number of return-to-work co‑ordinators is given in full-time equivalent
Source: OECD calculations based on Institut national d’assurance maladie‑invalidité – Baromètre Retour au Travail.
Partial return to work can be an attractive option for workers on sickness absence
For workers with health issues, partial return to work can enable an earlier re‑entry into the labour market while taking into account diminished capacity to work. In recent years, a number of OECD countries have established systems that allows for a part-time return to work following sickness absence with partial income replacement (OECD, 2023[56]).
In Belgium, the option for partial return to work during disability benefit receipt has existed since 1996, but has gained in popularity in recent years (De Brouwer and Tojerow, 2023[51]). Workers can take the partial return‑to-work option both during primary incapacity and long-term invalidity. Between 2017 and 2023, the number of people in long-term invalidity who have partial authorisation to work increased from 11.4% to 16.2% (Figure 2.9). Under the partial return-to-work system, workers can combine a remunerated activity – either adapted work with the previous employer or a new job – with partial benefit receipt (Conseil supérieur de l’emploi, 2024[52]). To access partial return to work, the GP must give approval, which can be given for a period of two years (permanently until 2018) and can subsequently be renewed. Partial return to work is possible based on the condition that work incapacity remains at least 50%.
Experience from other OECD countries shows that access to part-time employment can be an important facilitator of return to work after long-term sick leave, and the existence of such a possibility in Belgium is promising. Particularly against the background of increasing take‑up of partial return to scheme, more evidence on the impact of the scheme should be collected to investigate medium- and long-term outcomes and the impact of policy design choices. For instance, the role of financial incentives for shaping moves from inactivity into part-time work, but also part-time return to work to full-time employment, as well as distributional effects, should be further examined (Conseil supérieur de l’emploi, 2024[52]; Hufkens et al., 2016[59]).
Share of workers in long-term invalidity with authorisation for partial return to work, over time and by age
Note: Long-term invalidity (invalidité) refers to work capacity of one year or more.
Source: OECD calculations based on Institut national d’assurance maladie‑invalidité – Baromètre Retour au Travail.
References
[14] Adalet McGowan, M. et al. (2020), “Addressing labour market challenges in Belgium”, OECD Economics Department Working Papers, No. 1602, OECD Publishing, Paris, https://doi.org/10.1787/8a542d68-en.
[29] Albanese, A. and B. Cockx (2019), “Permanent wage cost subsidies for older workers. An effective tool for employment retention and postponing early retirement?”, Labour Economics, Vol. 58, pp. 145-166, https://doi.org/10.1016/j.labeco.2018.01.005.
[30] Albanese, A. and B. Cockx (2017), “Les allegements permanent des charges salariales pour travailleurs agés. Un instrument efficace pour maintenir l’emploi et retarder la prepension?”, Revue Belge de Sécurité Sociale, pp. 353-372.
[21] Baert, S. et al. (2016), “Getting grey hairs in the labour market. An alternative experiment on age discrimination”, Journal of Economic Psychology, Vol. 57, pp. 86-101, https://doi.org/10.1016/j.joep.2016.10.002.
[5] Behaghel, L., B. Crépon and B. Sédillot (2008), “The perverse effects of partial employment protection reform: The case of French older workers”, Journal of Public Economics, Vol. 92/3-4, pp. 696-721, https://doi.org/10.1016/j.jpubeco.2007.12.004.
[55] Boets, I. et al. (2020), “Evaluation de l’impact de la nouvelle réglementation sur la réintégration au travail: Synthèse de l’évaluation quantitative, qualitative et juridique”.
[54] Boets, I. et al. (2024), “Success rate and predictors of return to work after implementation of a formal return-to-work trajectory: A retrospective cohort study”, Work, Vol. 79/3, pp. 1147-1157, https://doi.org/10.3233/wor-230412.
[28] Boockmann, B. (2015), “The effects of wage subsidies for older workers”, IZA World of Labor, Vol. 189, https://doi.org/10.15185/izawol.189.
[37] Carpentier, M. et al. (2023), “Interregionale tewerkstellingszones op basis van een vraag- en aanbodanalyse”, No. VIONA Report D/2023/3241/171.
[36] Causa, O., M. Abendschein and M. Cavalleri (2021), “The laws of attraction: Economic drivers of inter-regional migration, housing costs and the role of policies”, OECD Economics Department Working Papers, No. 1679, OECD Publishing, Paris, https://doi.org/10.1787/da8e368a-en.
[2] Causa, O. et al. (2022), “Getting on the job ladder: The policy drivers of hiring transitions”, OECD Economics Department Working Papers, No. 1710, OECD Publishing, Paris, https://doi.org/10.1787/0304c673-en.
[33] Causa, O., N. Luu and M. Abendschein (2021), “Labour market transitions across OECD countries: Stylised facts”, OECD Economics Department Working Papers, No. 1692, OECD Publishing, Paris, https://doi.org/10.1787/62c85872-en.
[1] Cederlöf, J. et al. (2024), “Mandatory Notice of Layoff, Job Search, and Efficiency”, The Quarterly Journal of Economics, Vol. 140/1, pp. 585-633, https://doi.org/10.1093/qje/qjae029.
[12] Conseil Central de l’Economie (2020), “Rapport sur la rémunération en fonction de l’ancienneté”.
[52] Conseil supérieur de l’emploi (2024), “Incapacité de travail et réintégration des salariés sur le marché de l’emploi”, Brussels.
[13] Conseil supérieur de l’emploi (2014), “Avis sur la relation entre salaire et ancienneté”.
[64] Conseil Superieur des Finances (2025), “Comité d’Étude sur le Vieillissement - Rapport Annuel”, https://conseilsuperieurdesfinances.be/sites/default/files/publications/csf_vieillissement_2025_07.pdf (accessed on 17 December 2025).
[50] Cottier, L., K. Degen and R. Lalive (2019), “Can unemployment benefit cuts improve employment and earnings?”, Empirical Economics, Vol. 59/2, pp. 659-699, https://doi.org/10.1007/s00181-019-01677-6.
[7] D’Addio, A., M. Keese and E. Whitehouse (2010), “Population ageing and labour markets”, Oxford Review of Economic Policy, Vol. 26/4, pp. 613-635, https://doi.org/10.1093/oxrep/grq035.
[22] Dalle, A., L. Lippens and S. Baert (2024), “Nothing really matters: evaluating demand-side moderators of age discrimination in hiring”, Socio-Economic Review, Vol. 23/3, pp. 1253-1285, https://doi.org/10.1093/ser/mwae070.
[26] Dalle, A., E. Verhofstadt and S. Baert (2024), “The subsidy trap: Explaining the unsatisfactory effectiveness of hiring subsidies for the senior unemployed”, Journal of Economic Behavior & Organization, Vol. 227, p. 106713, https://doi.org/10.1016/j.jebo.2024.106713.
[38] De Brouwer, O. and I. Tojerow (2024), “Old-age unemployment and labour supply: an application to Belgium”, Empirical Economics, Vol. 67/1, pp. 253-287, https://doi.org/10.1007/s00181-023-02544-1.
[51] De Brouwer, O. and I. Tojerow (2023), “The Growth of Disability Insurance in Belgium: Determinants and Policy Implications”, IZA DP No. 16376.
[45] de Groot, N. and B. van der Klaauw (2019), “The effects of reducing the entitlement period to unemployment insurance benefits”, Labour Economics, Vol. 57, pp. 195-208, https://doi.org/10.1016/j.labeco.2019.02.003.
[24] Department Werk en Sociale Economie (2024), “Rapport Sectorale Nulmeting. Monitoringsoefening non-discriminatie 2021-2024”, Department Werk en Sociale Economie, Brussels.
[31] Desiere, S., S. Cabus and B. Cockx (2020), “Evaluatie van het Vlaamse doelgroepenbeleid”, HIVA KU Leuven.
[41] Desmette, D. and P. Vendramin (2021), “Bridge employment in Belgium: Between an early retirement culture and a concern for work sustainability”, in Alcover, C. et al. (eds.), Bridge Employment: A research handbook, Routledge, London.
[57] Dol, M. et al. (2021), “Systematic Review of the Impact on Return to Work of Return-to-Work Coordinators”, Journal of Occupational Rehabilitation, Vol. 31/4, pp. 675-698, https://doi.org/10.1007/s10926-021-09975-6.
[8] Eurofound (2019), Seniority-based entitlements: extent, policy debates and research.
[61] Eurofound (n.d.), A policy of the past? Seniority-based entitlements, http://www.eurofound.europa.eu.
[46] Filges, T., A. Jonassen and A. Jørgensen (2018), “Reducing unemployment benefit duration to increase job finding rates: a systematic review”, Campbell Systematic Reviews, Vol. 14/1, pp. 1-194, https://doi.org/10.4073/csr.2018.2.
[58] FPS Santé publique, S. (2025), HWF Statistiques Annuelles 2024, https://organesdeconcertation.sante.belgique.be/fr/documents/hwf-statistiques-annuelles-2024 (accessed on 10 April 2026).
[40] Fraikin, A., A. Jousten and M. Lefebvre (2021), “Social Security Incentives in Belgium: An Analysis of Four Decades of Change”, in Börsch-Supan, A. and C. Coile (eds.), : Social Security Programs and Retirement around the World: Reforms and Retirement Incentive, University of Chicago Press, Chicago.
[10] Frimmel, W. et al. (2018), “Seniority wages and the role of firms in retirement”, Journal of Public Economics, Vol. 164, pp. 19-32, https://doi.org/10.1016/j.jpubeco.2018.04.013.
[15] Gagliardi, N., E. Grinza and F. Rycx (2022), “Workers’ tenure and firm productivity: New evidence from matched employer‐employee panel data”, Industrial Relations: A Journal of Economy and Society, Vol. 62/1, pp. 3-33, https://doi.org/10.1111/irel.12309.
[63] Henkens, K. (ed.) (2012), “Ageing and employers’ perceptions of labour costs and productivity”, International Journal of Manpower, Vol. 33/6, pp. 629-647, https://doi.org/10.1108/01437721211261796.
[59] Hufkens, T. et al. (2016), “Fit for the labour market? An effort to reduce inactivity traps in the transition from benefit to work in the Belgian sickness and disability system”, Herman Deleeck Centre for Social Policy, University of Antwerp.
[27] Huttunen, K., J. Pirttilä and R. Uusitalo (2013), “The employment effects of low-wage subsidies”, Journal of Public Economics, Vol. 97, pp. 49-60, https://doi.org/10.1016/j.jpubeco.2012.09.007.
[60] INAMI (2026), Baromètre Retour au Travail pour les salariés, https://www.inami.fgov.be/fr/statistiques/statistiques-indemnites/barometre-retour-au-travail-pour-les-salaries (accessed on 16 January 2026).
[62] Institute for Applied Economic Research at the University of Tübingen, Germany (2015), “The effects of wage subsidies for older workers”, IZA World of Labor, https://doi.org/10.15185/izawol.189.
[47] Kyyrä, T. and H. Pesola (2020), “The effects of unemployment benefit duration: Evidence from residual benefit duration”, Labour Economics, Vol. 65, p. 101859, https://doi.org/10.1016/j.labeco.2020.101859.
[44] Le Barbanchon, T. (2016), “The effect of the potential duration of unemployment benefits on unemployment exits to work and match quality in France”, Labour Economics, Vol. 42, pp. 16-29, https://doi.org/10.1016/j.labeco.2016.06.003.
[17] Lippens, L., S. Vermeiren and S. Baert (2023), “The state of hiring discrimination: A meta-analysis of (almost) all recent correspondence experiments”, European Economic Review, Vol. 151, p. 104315, https://doi.org/10.1016/j.euroecorev.2022.104315.
[11] Minne, G. and Y. Saks (2023), “Older workers and retirement decisions in Belgium: mapping insights from survey data”, NBB Economic Review, Vol. 15.
[48] Nekoei, A. and A. Weber (2017), “Does Extending Unemployment Benefits Improve Job Quality?”, American Economic Review, Vol. 107/2, pp. 527-561, https://doi.org/10.1257/aer.20150528.
[32] OECD (2025), Addressing Regional Labour Market Imbalances in Austria, OECD Reviews on Local Job Creation, OECD Publishing, Paris, https://doi.org/10.1787/0cf6186b-en.
[42] OECD (2025), Pensions at a Glance 2025: OECD and G20 Indicators, OECD Publishing, Paris, https://doi.org/10.1787/e40274c1-en.
[39] OECD (2024), OECD Economic Surveys: Belgium 2024, OECD Publishing, Paris, https://doi.org/10.1787/c671124e-en.
[6] OECD (2024), Promoting Better Career Choices for Longer Working Lives: Stepping Up Not Stepping Out, Ageing and Employment Policies, OECD Publishing, Paris, https://doi.org/10.1787/1ef9a0d0-en.
[56] OECD (2023), Disability, Work and Inclusion in Korea: Towards Equitable and Adequate Social Protection for Sick Workers, OECD Publishing, Paris, https://doi.org/10.1787/bf947f82-en.
[3] OECD (2020), OECD Employment Outlook 2020: Worker Security and the COVID-19 Crisis, OECD Publishing, Paris, https://doi.org/10.1787/1686c758-en.
[25] OECD (2019), Working Better with Age, Ageing and Employment Policies, OECD Publishing, Paris, https://doi.org/10.1787/c4d4f66a-en.
[53] OECD (2015), Fit Mind, Fit Job: From Evidence to Practice in Mental Health and Work, Mental Health and Work, OECD Publishing, Paris, https://doi.org/10.1787/9789264228283-en.
[18] OECD/Generation: You Employed, Inc. (2023), The Midcareer Opportunity: Meeting the Challenges of an Ageing Workforce, OECD Publishing, Paris, https://doi.org/10.1787/ed91b0c7-en.
[16] Saks, Y. (2014), “Salariés: trop chers à 50 ans? La composante «âge» dans la formation des salaires”.
[49] Schmieder, J., T. von Wachter and S. Bender (2016), “The Effect of Unemployment Benefits and Nonemployment Durations on Wages”, American Economic Review, Vol. 106/3, pp. 739-777, https://doi.org/10.1257/aer.20141566.
[43] Sjöberg, O. (2007), “Labour market mobility and workers’ skills in a comparative perspective: exploring the role of unemployment insurance benefits”, International Journal of Social Welfare, Vol. 17/1, pp. 74-83, https://doi.org/10.1111/j.1468-2397.2007.00493.x.
[34] SPF Emploi, T. (2022), “Etat des lieux de la mobilité professionnelle en Belgique”.
[35] Steunpunt Werk (2025), “Vlaamse Arbeidsrekening o.b.v. DWH AM&SB - KSZ, BISA”, https://www.steunpuntwerk.be/cijfers/vlaamse-arbeidsrekening (accessed on 30 March 2026).
[23] Tobback, I. et al. (2024), “Discrimination sur le marché du travail bruxellois : analyse sectorielle basée sur un test par correspondance”, No. Etude commandée par le Service public régional de Bruxelles (SPRB/GOB).
[20] UNIA (2025), “Research on age discrimination in Belgium”.
[19] Van Borm, H., I. Burn and S. Baert (2021), “What Does a Job Candidate’s Age Signal to Employers?”, Labour Economics, Vol. 71, p. 102003, https://doi.org/10.1016/j.labeco.2021.102003.
[4] Vodopivec, M. et al. (2019), “Increasing Employment of Older Workers: Addressing Labour Market Obstacles”, Journal of Population Ageing, Vol. 12/3, pp. 273-298, https://doi.org/10.1007/s12062-018-9236-4.
[9] Zwick, T. (2012), “Consequences of Seniority Wages on the Employment Structure”, ILR Review, Vol. 65/1, pp. 108-125, https://doi.org/10.1177/001979391206500106.